effect of information asymmetry on borrowing cost …

International Journal of Economics, Commerce and Management United Kingdom ISSN 2348 0386 Vol. VII, Issue 8, August 2019

Licensed under Creative Common Page 440

http://ijecm.co.uk/

EFFECT OF INFORMATION ASYMMETRY ON BORROWING

COST AMONG MICRO-FINANCE CLIENTS IN KENYA

E. N. Muli

Certified Public Accountant, Kenya

[email protected]

W. M. Muli

University Lecturer, Kenya

[email protected]

Abstract

Money and credit are the lifeblood of an economy. The ability of borrowers to access credit at

reasonable terms is critical to facilitate investment and commerce, and thereby sustain

economic growth. The poor still find it difficult to access finances from MFIs because of the fairly

high cost of loans. Most microfinance clients are information opaque and this partly explains the

high risk premium attached to lending hence higher interest rates. Accordingly, the purpose of

this study was to find out the effect of information asymmetry on the cost of borrowing among

microfinance clients. Specifically, the study sought to find out: effect of credit history on the cost

of borrowing; effect of soft information on the cost of borrowing; and, effect of borrower proximity

on the cost of borrowing. The study employed a descriptive survey design. This research relied

purely on primary data which was collected using a structured questionnaire. Both descriptive

and inferential statistics were used for data analysis. A multiple regression model was used to

estimate the relationship between information asymmetry and cost of borrowing. The study

findings show that borrower credit history and soft information are significant influencers of

borrowing cost among microfinance borrowers in Kenya. Favourable borrower credit history and

soft information have a favourable effect of reducing the borrower cost of borrowing among

microfinance borrowers in Kenya. However, the study also concludes that in the Kenyan context

borrower proximity has no effect on the cost of borrowing among microfinance borrowers in

Kenya. This implies that microfinance institutions in Kenya do not take into account the physical

http://ijecm.co.uk/

http://ijecm.co.uk/

International Journal of Economics, Commerce and Management, United Kingdom


distance of the borrower from the MFI when evaluating a loan application. The study

recommends that microfinance institutions in Kenya should leverage borrower credit history and

soft information to mitigate information asymmetry challenges. This practice will also be

beneficial to the borrowers who possess good credit history standing and positive soft

information. By extension this will make the bottom of the society (microfinance clients) to

access credit at favourable terms. In turn, this will have a positive impact on the economy by

mainstreaming the poor into economic participation through affordable credit which can be used

to finance business operations sustainably.

Keywords: Information asymmetry, Borrowing cost, Microfinance clients, Kenya

INTRODUCTION

The role of information asymmetry in debt contracting has long been of interest to researchers

in accounting and finance. Information acquisition by financial intermediaries is an essential

function. It can improve the allocation of credit in the economy, and it is one of the main sources

of bank profits. Better knowledge of their loan applicants allows banks to weed out low-quality

projects.

The concept of asymmetric information was first introduced by Akerlof (1970) in a paper

entitled The Market for “lemons”. In the paper, Akerlof (1970) developed asymmetric information

with the example case of automobile market. His basic argument was that in many markets the

buyer uses some market statistic to measure the value of a class of goods. Thus the buyer sees

the average of the whole market while the seller has more intimate knowledge of a specific item.

Akerlof (1970) argued that this information asymmetry gives the seller an incentive to sell goods

of less than the average market quality.

In credit markets, asymmetric information problems arise when borrowers have private

information about their creditworthiness that is not observable by lenders. In general, the price

of a loan is based on the lender‟s cost plus a risk premium. The cost of funds is often linked to a

short-term market rate, which represents a common benchmark for all borrowers regardless of

their credit history. Lenders will often charge an additional risk premium over the market rate as

compensation for bearing the risk of slow, partial, or fully delinquent loan repayments. Some

losses are expected in any risk group of loans and are in effect, paid for by the risk premium.

The size of this premium depends on the lender‟s ability to properly assess the creditworthiness

of the borrower. As a result, differences across borrowers in the final interest rate charged are

based upon largely the lender‟s perceived repayment risk (Leitner, 2006).

http://ijecm.co.uk/

© Muli & Muli


When lenders lack the necessary information to distinguish between good and bad borrowers, it

is said that there exist “adverse selection” and “moral hazard” problems in the market for credit.

Adverse selection occurs when a borrower‟s private information about their own credit risk

adversely affects uninformed lenders. For instance, with limited credit information, there are

likely to be more bad borrowers taking loans at any given interest rate. Moral hazard entails

hidden information following the extension of a loan to a borrower. For instance, if a borrower

knows that a lender cannot monitor repayment behaviour, this can induce the borrower to make

a material change in income or spending that affects their ability to repay the loan (Jappelli and

Pagano, 2002).

Several studies have been conducted across the world on the subject of information

asymmetry and borrowing. For instance in Portugal, Bonfim, Dai and Franco (2009) provided

evidence on the effect of bank competition on the cost of lending, in an environment of reduced

information asymmetries between firms and banks. The authors constructed a simple model

linking the number of bank relationships, the cost of lending and bank competition. Banks are

exposed to more competition if the firm has many ongoing bank relationships that improve her

threat point when negotiating borrowing costs. Moreover, increased competition in the banking

sector might mitigate (substitute) or amplify (complement) this effect. Using a unique data set

from Portugal, they find that when a firm borrows from one additional bank, the interest rate on

bank loans for this firm becomes 9 to 20 basis points lower on average. In addition, it is found

that when local bank competition is more intense firms can benefit more from simultaneously

engaging in several banking relationships, hence providing evidence of complementarity

between competition and the number of bank relationships. However, these effects are not

observed for the smallest and youngest firms.

Using panel data Taiwan and Huang (2014) examined the impact of information

asymmetry and client credit on lending performance. The empirical results showed that: First,

MLBs with good information transparency tend to establish relationships with banks that are

characterized by huge losses from bad debts and from credit card lending. Second, Small

foreign firms, as well as MLBs with high profitability, cash and R and D expenditure ratios prefer

having relationships with banks with good lending performance and low credit risk. Third, MLBs

and MSBs with poor credit records prefer having relationships with banks that have good

lending performance and low credit risk.

Bolton, Freixas, Gambacorta and Mistrulli (2013), studied how relationship lending and

transaction lending vary over the business cycle. They were able to study how relationship and

transaction-banks responded to the crisis and tested existing theories of relationship banking.

The empirical results established that relationship banks charged a higher spread before the


Licensed under Creative Common 43

crisis, offered more favourable continuation-lending terms in response to the crisis, and suffered

fewer defaults, thus confirming the informational advantage of relationship banking.

In Kenya, Gaitho (2013) explored the effect of credit reference bureaus (CRBs) on credit

access. The study found out that CRB reduces borrowing cost and loan delinquencies to a

moderate extent. It further established that CRB has enhanced effective risk

identification/monitoring and microcredit extension in Kenya. It was therefore recommended that

lenders and CRB should work closely to ensure that there is no information asymmetry and

therefore ensure that credit flows to deserving borrowers.

In another Kenyan study, Kwambai and Wandera (2013) sought to find out the effects of

credit information sharing on nonperforming loans in KCB Kenya and specifically to establish

the trend of bad loans before and after the introduction of CRB, to identify the factors that

account for bad loans and to determine the economic sector that records higher bad loans and

the efforts taken to reduce the risk in this sector. The study concluded that credit information

sharing and level of nonperforming loans are indeed related. Credit Information Sharing,

increases transparency among financial institutions, helps the banks lend prudently, lowers the

risk level to the banks, acts as a borrowers discipline against defaulting and it also reduces the

borrowing cost, that is, interest charged on loans.

Concept of Information Asymmetry

According to Akerlof (1970) information asymmetry is a situation whereby the buyer in a market

sees the average of the whole market as the basis of evaluating quality of a product while on

the other hand the seller has more intimate knowledge of a specific item. Therefore, the seller is

at an advantageous position because he/she has superior information concerning quality of the

product as compared to the buyer. This skewed possession of information is referred to as

information asymmetry. According to Leitner (2006) asymmetric information is a situation

whereby borrowers have private information about their creditworthiness that is not observable

by lenders. Jappelli and Pagano (2002) wrote that when lenders lack the necessary information

to distinguish between good and bad borrowers, it is said that there exist “adverse selection”

and “moral hazard” problems in the market for credit. For the purpose of this study the Leitner

(2006) definition shall be adopted.

Information asymmetry can be measured using different parameters; Credit history, soft

information and borrower proximity. Borrower credit history is typically compiled by credit

reference bureaus. The purpose of the bureau is to collect, collate, analyze, and disseminate

credit information among credit providers. Credit information sharing provides credit history

(information capital) as an alternative form of collateral to the traditional physical collateral, to

http://ijecm.co.uk/

© Muli & Muli


secure credit facilities from banks. On the other hand, the bank benefits from the mechanism,

since it will address the problem of information asymmetry that is typically used to raise a risk

premium on loans. The problems of moral hazard and adverse selection are also minimized.

Credit histories not only provide necessary input for credit underwriting, but also allow borrowers

to take their credit history from one financial institution to another, thereby making lending

markets more competitive and, in the end, more affordable (Muli, 2013).

An evaluation of soft information could also reduce information asymmetry. Typically,

soft information is not directly verifiable by a third party and includes hypothetical and/or

intangible information such as economic projections, assessments of a firm‟s management

quality, and employee morale. Soft information is typically thought to be accessible exclusively

to a primary incumbent. Therefore, this information is only accessible in the context of

relationship banking (Campbell and Loumioto, 2013). Soft information can be interpreted as a

private signal about the quality of a firm that is observable to a relationship bank, but not to a

transaction bank. Previous research in relationship banking generally suggests that soft

information can improve contracting efficiency and affect a firm‟s costs and/or its access to

credit. Relationship lending primarily as a practice can be able to alleviate information

asymmetry. It lowers information asymmetries and monitoring costs, in a competitive loan

market, banks transfer monitoring cost savings to their clients in terms of lower interest rates. In

particular, it is expected that reduced information asymmetries through inside information

collected during face-to-face meetings are passed on to the borrower by means of a lower

interest rate (Salvatore, Capasso and Gabriele, 2015).

Borrower proximity plays a role in determining the level of information asymmetry.

According to Agarwal, Chomsisengphet and Souleles (2009) distance erodes lender‟s ability to

collect proprietary intelligence and to carve out local captive markets suggesting that the

requisite soft information is primarily local. Borrower proximity facilitates the collection of soft

information, thereby enhancing its quality, which is consistent with the notion that banks enjoy a

local informational advantage that distance erodes. Proximity facilitates the collection of soft

information, leading to a trade-off in the availability and pricing of credit, which is more readily

accessible to nearby firms albeit at higher interest rates ceteris paribus.

From the foregoing, for the purposes of this study, information asymmetry shall be

measured using three indicators of credit history (Muli, 2013; Jonathan and Jonathan, 2001;

Howorth and Moro, 2012); soft information (Campbell and Loumioto, 2013; Wanzare, 2010;

Sseggujja, 2010) and borrower proximity (Agarwal, Chomsisengphet and Souleles, 2009; Dass

and Mass, 2009).



Microfinance refers to small-scale financial services, primarily credit, savings and insurance.

Microfinance offers poor people and micro enterprises access to basic financial services such

as loans and savings. The information asymmetry is most pronounced in this borrower segment

because of lack of guarantees, lack of information (short credit history) and difficulty in

monitoring. In addition, these micro enterprises frequently do not possess legal documentation,

properties, or regularized wages, which are demanded by traditional banking institutions

(Hercules, 2006). Therefore, it would be correct to say that the bulk of microfinance clients are

information opaque.

The microfinance institutions have taken different measures to mitigate the information

asymmetry problem which is uniquely more pronounced among their clients. One of the key

innovations towards this end is group based lending. Group-based lending, as the term already

indicates, requires individuals to organize themselves into groups in order to gain access to

financial services from a MFI. Normally, group-based lending works as follows. Loans are made

to individuals, but all members of the group are held responsible for the loan repayment (joint

liability principle). In some programs loans are given strictly for a certain period of time (usually

a year), while in other programs the members are allowed to decide the loan terms themselves.

Repayments are made on a weekly or monthly basis; this is done at group meetings or directly

to the branches of the microfinance institution. Thus, in group-lending programs the functions of

screening, monitoring and enforcing repayments is to a large extent transferred from the bank

agent to group members. Peer pressure is a mechanism group members can use in the process

of mitigating moral hazard and enforcing punctual repayment. In order to secure future access,

members are obliged to monitor each other. Once output is realized and a member proves

unwilling to repay, other members can use peer pressure and social sanctions to make him

repay (Mehrteab, 2005)

Leveraging customer credit history information has also been used by MFIs to mitigate

losses occasioned by information asymmetry challenges. Various theories have explored how

sharing credit information can alleviate moral hazard, adverse selection and over-borrowing.

Sharing this information helps lenders avoid low-quality borrowers and in turn incentivizes

borrowers to stay off the blacklist. When lenders share positive information borrowers can

gradually build up a valuable reputation as trustworthy borrowers (Jaap, Ralph & Matteo, 2015).

If a potential client has had a bad credit record in the past, it is right for a bank to deal with such

a case in a prudent manner. While some banks may abandon this type of business to avoid a

high non-performing loan ration, other banks are attracted by a high loan spread, all of which

depends on the banks credit policy.

http://ijecm.co.uk/

© Muli & Muli


Additionally, Emilio, Galariotis, Villa & Yusupov (2011) opine that the problem of information

asymmetry among the low income borrowers and micro enterprises could be solved through

collateral requirement. A major part of the informational problem could be avoided in theory,

through the requirement of tangible collateral prior to lending such as furniture, livestock, and

personal effects.

Borrowing Cost

In the strict sense of finance, cost of borrowing is that amount which is paid to the providers of

capital. It either appears as a charge against the business to arrive at the profits for instance

interest on debentures or it appears as form distribution of the profits e.g. dividends on

preference shares and ordinary shares. The interest and the dividends both represent the cost

of obtaining and using capital (Pandey, 2005).

In this study, the cost of borrowing would be defined more broadly using the parameters

discussed below; higher interest (Bonfim, Dai and Franco, 2009), collateral requirement

(Leinter, 2006; Stiglitz and Weiss 1981), shorter debt maturity (Moerman, 2010) and credit

rationing (Amir, 2007). Interest charged on borrowed funds is the most visible element in the

cost of borrowing. Bonfim, Dai and Franco (2009) wrote that the impact of information

asymmetry on the cost of debt capital is that it influences the interest rate charged on the

borrower. A higher bid-ask spread will apply to the borrowers the lender does not know much

about. While on the other hand, borrowers with good credit history would enjoy lower interest

rates. Information asymmetry is priced in terms of interest rate.

Collateral is an important element in the cost of borrowing. Most lenders normally

demand for collateral in order to mitigate against information asymmetry challenges. Generally,

the term collateral refers to assets pledged by a borrower to secure a loan. The lender can seize

these assets if the borrower does not make the agreed-upon payments on the loan, so the

lender has some protection if the borrower defaults. Therefore, the use of collateral can make it

easier for borrowers to obtain loans to finance their investments (Leinter, 2006). Collateral is a

mechanism that mitigates adverse selection, credit rationing, and other inefficiencies that arise

when borrowers have ex ante private information. Stiglitz and Weiss (1981) explains collateral

as arising from ex ante information gaps between borrowers and lenders. Specifically, when

borrowers have private information regarding their project quality, the equilibrium may be

characterized by adverse selection and credit rationing. The pledging of collateral may allow

lenders to sort observationally equivalent loan applicants and mitigate these inefficiencies.

Debt maturity is also determined by information asymmetry. Moerman (2010) found out

that information asymmetry decreases debt maturity. Furthermore, it was established that a



higher bid-ask spread on the borrower‟s traded loans translates into a shorter maturity of the

borrower‟s subsequently issued loans. Therefore, borrowers with good credit history will be

allowed longer repayment period as compared to information opaque borrowers.

Credit rationing is another measure used by lenders to mitigate information asymmetry

challenges. Broadly speaking, „credit rationing‟ refers to any situation in which lenders are

unwilling to advance additional funds to a borrower even at a higher interest rate. It is a situation

in which the demand for commercial loans exceeds the supply of these loans at the commercial

loan rate quoted by the banks. The lender would deny credit to potential information opaque

borrowers in favour of familiar borrowers (Amir, 2007).

There are various ways to mitigate information asymmetry problems in the market for

credit. These measures either directly or indirectly constitute a cost to the borrower. For

example, collateral can be required for a loan; the lender can add risk premium on the lending

rates; the lender can ration credit when there is insufficient information about the borrower; and

the lender can demand shorter repayment period. In this study these four will be used as the

measures of the cost of borrowing.

Microfinance Institutions in Kenya

The microfinance sector in Kenya is fairly well developed by the Sub-Saharan Africa standards.

It constitutes of a variety of organisations whose objective is to reach the bottom of the society

with varied financial products. The Microfinance Act was formulated in the year 2006 to

specifically regulate this sector. Previously there was no piece of regulation relevant to this

sector. This led to a proliferation of a multiplicity of institutions and products aimed at this

hitherto unbanked market segment. The microfinance sector in Kenya has largely developed

under this scenario (Darko, 2013)

K-rep was the pioneer microfinance institution in Kenya. It was established in the year 1984

as a non-governmental organization (NGO) before evolving into a microfinance institution (MFI).

The conversion from NGO into MFI was motivated by the need to enhance capacity to mobilise

savings from the public for onward lending. The records kept by the Association of Microfinance

Institutions (AMFI) indicate that today we have 49 registered MFIs in Kenya.

Statement of Research Problem

According to Sabana (2003) microfinance is very critical in any economy because it helps in

mainstreaming the poor into economic participation through provision of credit and other

financial services. However, the poor still find it difficult to access finances from MFIs because

of the fairly high cost of loans. Most microfinance clients are information opaque and this partly

http://ijecm.co.uk/

© Muli & Muli


explains the high risk premium attached to lending hence higher interest rates (Thomas and

Worrall, 2000).

The provision of microcredits to poor people has been shown to help people work their

own way out of poverty. In fact, Microcredits have become a prime component of development

strategy worldwide, and the UN nominated year 2005 to be the year of microcredit. However,

there are many factors to be considered in order to develop a sustainable microcredit program

reaching poor. Despite being hailed as the refuge for the poor, MFIs have been accused of

levying higher interest rates as compared to the mainstream commercial banks besides

demanding collateral and shorter maturity periods. The high interest rates could be motivated by

the fact that there are high cost associated with screening the information opaque loan

applicants and enforcing repayment (Bellucci, Borisov & Zazzaro, 2013).

Therefore, it is critical that a study be carried out to identify and measure the entire

borrowing cost imposed by the MFIs on the poor borrowers and micro enterprises. In order to

operate their businesses sustainably, the borrowing cost must be reasonably affordable. The

study will offer insights and recommendations on how MFIs could lower the borrowing cost in

the face of information asymmetry. No previous studies have attempted to measure the entirety

of this cost and offer suggestions for a reduction.

Studies across the world have been undertaken to evaluate the effect of information

asymmetry on the cost of borrowing. Moerman (2010) evaluated the impact of information

asymmetry on debt pricing (interest rate) and maturity. This study relied on only a single

measure of information asymmetry namely credit history; Bonfim, Dai and Franco (2009)

investigated the impact of information asymmetry on the cost of debt capital (interest rate). In

this study the author relied on only one measure of borrowing cost namely interest rates; Leinter

(2006) evaluated the influence of information asymmetry on collateral requirements. The

researcher measured the cost of borrowing using collateral requirement only; Amir (2007)

analyzed the effect of information asymmetry on credit rationing. In this study the cost of

borrowing was measured using only one indicator – credit ration; In Kenyan Kwambai and

Wandera (2013) sought to find out the effects of credit information sharing on nonperforming

loans in KCB Kenya. The study evaluated the cost of information asymmetry to the lender as

opposed to the borrowers; in another study Gaitho (2013) explored the effect of credit reference

bureaus (CRBs) on credit access in Kenya. In this study information asymmetry was purely

measured using credit history.

All the different measures aimed at protecting the lender against information asymmetry

namely higher interest, collateral requirement, shorter debt maturity and credit rationing

constitute a cost to the borrower. No empirical study has tried to comprehensively measure the



cost of information asymmetry to the borrowers of funds. Most studies done previously have

only focused on the cost the borrower suffers as a result of a risk premium being added on the

interest rate charged (hence the borrower being charged a higher interest rate). This study

intended to broadly capture all indicators of cost to the borrower including: requirement for

collateral, shorter debt maturity, credit rationing, and higher interest rate. The study sought to

answer the question: what is the effect of information asymmetry on the cost of borrowing

among microfinance clients?

RELATED LITERATURE

In this section, previous related studies are evaluated. The discussion focuses on the study

objectives, context, methodologies employed, and the findings. But most important the

knowledge gaps in those studies are brought out.

Credit History and the Borrowing Cost

Employing information asymmetry indicators based on the World Bank‟s Doing Business

survey, Barbosa and Marcal (2011) investigated the role that information asymmetry plays in

bank spreads. The dependent variable was a country‟s average bank spread, selected to

measure the cost of bank loans in the credit market, independent variables included information

rate, degree of coverage of the population of public credit bureaus and degree of coverage of

the population of private credit bureaus. Control variables used included the country‟s tax

burden, equity capital requirements, inflation, default level and quality of the legal system. The

number of countries with prime rate data information recorded in the IMF and World Bank

databases for the years under study defined the sample size. Based on a universe of 196

countries with a total of 980 notes, a sample was selected containing 14 countries with valid

data, with recorded loan rates intended for a lower-risk public (prime rates) with 70 notes from

2002 to 2006. The results showed that the existence of a lesser degree of information

asymmetry in credit markets reduces bank spreads. This conclusion was obtained based on a

study of prime loan rates. The effect would be between a 2% to 4% permanent reduction in

spreads. The relation was obtained based on econometric panel data models with static

effects. This study only used one measure of borrowing cost – bank spreads (interest rates).

The current study will measure cost of borrowing more broadly.

In Kenya, Ngugi (2001) sought to explain the factors determining interest rate spread for

Kenya‟s banking sector. In the data it was assumed that interest rate depends on demand for

investment funds and other composite variables including the alternative sources. The data

http://ijecm.co.uk/

© Muli & Muli


consisted of monthly observations of Treasury bill rates, commercial bank loans and deposits,

lending rates, inter-bank rates, provision for bad loans, and liquidity and cash ratios. These data

were obtained from the Central Bank of Kenya. The sample ran from July 1991 to December

1999 for all data set except the inter-bank rate, which was only available from April 1993. A

regression model was run to establish the relationships. The results indicated that banks

factored information asymmetry risk in pricing the loans. To cover credit risk, banks charge a

premium whose size depends on the bank credit policy. Customers with good credit history

normally enjoy lower interest rates and vice versa. This study evaluated interest rate as the only

indicator of borrowing cost.

In China, Huang et al. (2014) conducted a narrative literature review using financing

theories of SMEs information asymmetry and credit rationing theories to analyze the financing

difficulties for SMEs. The researchers concluded that information asymmetry leads to adverse

selection and moral hazard and hence higher interest rates. This makes a large number of

SMEs to have no access to loans since banks are reluctant to lend. There is increased

transaction cost associated with information search in a situation of information asymmetry.

Where credit history information is not available, the bank has to take efforts in terms of money

and time to search for information which can reflect the real situation of the enterprise to ensure

the safety and efficiency of loans and prevent borrower default. These information varying from

financial statements, the credit rating level to quality of management, cash flow, business

prospects, and so forth. However, information related to this is not easy to obtain in the credit

market, because most of it is highly internalized and not accessible, which will drive up the cost

of information searching. This cost is factored in determination of the interest rates.

Howorth and Moro (2012) examined whether lending manager‟s assessments of the

trustworthiness of SME owner managers are associated with interest rate charged. Data were

obtained from a survey of lending managers from small banks in North East Italy. The

dependent variable was interest rates paid by each SME on their overdraft as reported by the

bank lending manager. The independent variable was trust which was measured using the

manager‟s ability to manage business, benevolence and integrity. Market power of the bank

was used as a control variable. The data was collected on a random sample of customers,

resulting in 365 small firms representing a 74% response rate. Multivariate regression analysis

found evidence of a negative association between trustworthiness and interest rates. This study

purely used interest rate as the cost of borrowing.

Jonathan and Jonathan (2001) examined the proposition that informal forms of finance

might play a significant role in overcoming both information asymmetry and the finance gap. The

aims of the research were achieved by means of a two-stage approach. Firstly, grounding



interviews were conducted with providers and recipients of finance, as well as advisors and

intermediaries in the small business finance market. Secondly, 1 000 small UK firms were

surveyed to investigate the patterns in financing as well as difficulties encountered in the

financing process. The exploratory nature of this initial research phase favored a methodological

approach that was inductive in nature. Consequently, a qualitative, interview-based approach

was adopted. The study concluded that the loan provider can reduce credit risk by carefully

screening firms at the outset and monitoring projects during the life of the loan. However,

screening and monitoring are high cost activities associated with the lending proposition. If the

lender is to recoup these costs then borrower interest rates may be increased, additional risk

may be covered by demanding collateral or may be avoided altogether by rejecting the loan

application. To summarize, the general problem of information asymmetry can manifest itself in

one of three ways: acceptance of the loan application but at a higher than risk-adjusted interest

rate; acceptance but with strict collateral requirements; or outright rejection of the loan

application. The study identified the measures taken by lenders to resolve information

asymmetry problems but did not measure the total cost occasioned by these measures to the

borrower.

Soft Information and Cost of Borrowing

Campbell and Loumioto (2013) studied the portability of soft information in a decentralized

financial institution. Using unique data on lending decisions made by employees in a highly

decentralized financial services organization, the authors showed that a monitoring system

which captures soft information for vertical communication (to superiors) purposes also

facilitates the horizontal communication of soft information (across employees) for decision

making purposes. The research site for the study was a federal credit-union with approximately

$1.6 billion in assets and 140 000 customers, and 23 branches operating in a single state in the

US. The primary data for the study came from the organizations internal lending, personnel, and

customer records during the period 2008 – 2010. Throughout this period, this organization when

compared to a peer group of same-state credit unions or to a national peer group of similar size,

has consistently ranked in the top 15% in productivity, loan default rates (2nd lowest), and overall

performance (return –on- assets). The study employed an OLS model in the analysis. The

results provide evidence that the “stock” of soft information accumulated in this system has

persistent effects on the lending decisions of employees. Employees rely on this information to

increase access to credit for borrowers, provide more favorable pricing terms, and reduce the ex

post risk of their lending decisions.

http://ijecm.co.uk/

© Muli & Muli


Wanzare (2010) used US loan-level data set between 2000 - 2015 identified soft information

and its impact on borrower default. Soft information was described as information that is hard to

express numerically and is only known by the lender exposed to it. The study analyzed the

impact of soft information on the mortgage rate and its effect on borrower default. A regression

model was used in the analysis. The study finds out that soft information is a positive and

significant predictor of borrower default and that soft information gathered by lenders is an input

into the pricing of the loan.

Aurizion, et al. (2012) studied how access to bank lending during financial crisis differed

between family and non-family firms. The theoretical underpinning of the study was that the

presence of a family block-holder in the company attenuates the agency conflict in the borrower-

lender relation, because of higher non-monetary costs of default entailed in this type of

corporate ownership structure. Because this information is to a large extent soft, the

researchers further investigated the interaction between the family firm status and the screening

technology adopted by the banks. The dependent variable was the difference in net interest rate

charged and the time difference in the collateral-ratio before and after the crisis. Dummy

variable distinguishing family and non-family firms was used as the independent variable. The

study finds that family firms experienced a contraction in credit granted than non-family firms.

The results are robust to ex-ante differences between the two types of firms and to bank-

specific firms. Banks that increased the role of soft information in their lending practiced

reallocated credited towards family firms.

In Uganda, Sseggujja (2010) investigated the relationship between relationship lending,

transaction costs and lending interest rates. Relationship lending was measured using: multiple

banking relationships, duration of the relationships, pre-existing relationship and trust. Primary

data was collected from 14 commercial banks in Uganda and their borrowers. A sample of 225

was drawn from the population of 566 medium and large sized borrowing enterprises and bank

employees of credit departments. Regression model was the tool of analysis. The study findings

reveal that relationship lending has a significant negative effect on lending interest rates and

transaction costs. Therefore, the study concludes that relationship lending and transaction costs

have a role in commercial bank pricing and loan application acceptance decision making.

Relationship lending is theoretically expected to reduce asymmetric information, which

potentially creates benefits for borrowers. However, empirical evidence is mixed. This motivated

Kysucky and Norden (2013) to conduct a meta-analysis to summarize and explain the

heterogeneity in the results in the literature using hand-collected information from 101 studies in

the U.S., Europe, Asia and Latin America from 1970-2010. The study finds that strong

relationships are generally beneficial for borrowers but lending outcomes differ across the



relationships‟ dimensions. Long-lasting, exclusive and synergy-creating bank relationships are

associated with higher credit volume and lower loan rates. These benefits are more likely in the

United States of America and in countries where bank competition is high. The higher the

deposits to GDP ratio, the higher the importance of SMEs in an economy; suggesting that a

higher prevalence if relationship lending does not necessarily come along with higher benefits

for borrowers.

Borrower Proximity and Cost of Borrowing

Bellucci et al. (2013) explored the effects of bank-borrower physical proximity on price and non-

price aspects of small business lending in local credit markets in Italy. The study used price and

availability as the dependent variables. The independent variable was the distance between the

bank and each borrower. The study also used a broad set of control variable reflecting borrower

characteristics and the nature of bank-borrower interaction. Relationships were estimated using

ordinary least squares method. The data used was collected over the period 2004 – 2006.

Along the price dimension, the analysis revealed that interest rates increase with bank-borrower

distance and decrease with the distance between bank and other competing banks. Along the

quantity dimension, it was observed that more distant borrowers are more likely to experience

binding credit limits. Results also showed that quantity effects of bank-borrower distance are

concentrated among less transparent firms.

Commercial banks acquire inside information about the firms they led to. Dass and Mass

(2009) studied the impact of this informational privilege position on the borrowing firm using a

broad panel of U.S. firms over the 1993 – 2004 period. The strength of bank-firm relationship

was measured by bank-firm proximity, size of the loan, and the lender‟s insider potential.

Proximity was defined as the geographical proximity between the borrower and the lender, the

significance of the loan to the borrowing firm‟s finances was measured using the loan-to-asset

ratio, and the bank‟s insider potential was measured by the equity ownership of the bank in the

borrowing firm. A cross-sectional econometric model was employed in the study. It was found

out that simultaneously; proximity and loan significance should benefit the borrower‟s

governance. Proximity directly affects the bank‟s information-gathering and monitoring ability

and hence the interest rates. The results showed that a stronger relationship, by inducing better

monitoring, improves borrower‟s corporate governance. Simultaneously, it makes the bank a

potentially more informed agent in the equity market. This information asymmetry increases

adverse selection for the other market participants and lowers the firm‟s stock liquidity. This

trade-off between improved corporate governance and greater information asymmetry affects

the firm‟s value.

http://ijecm.co.uk/

© Muli & Muli


Ostromogolsky (2017) conducted a study on the relationship between the borrowing choices

made by small businesses and their loan origination costs in the United States. The researcher

made a comparison between firms that borrow from a previously unused financial institution with

firms that borrow from a financial institution with which they have a pre-existing financial

relationship. The study assessed the causal effect of borrowing from a new lender on small

businesses‟ up-front closing costs. Loan origination cost was defined as the total dollar amount

of fees paid by a borrower to apply for and obtain a loan at the time of origination. Simple cross

sectional regression showed that small businesses that turn to new financial institution paid $5

650 to 6 740 more in closing costs than firms that return to a previously used institution. The

study also studied a natural quasi-experiment wherein the treatment of borrowing from a new

financial institution was close to was randomly assigned. A unique group of small firms

businesses that selected a lender based on the lender‟s distance to the firm was considered.

Data of firms that select a lender based on proximity revealed that borrowing from a new

financial institution significantly raises a firm‟s loan origination costs.

Knowledge Gap

An evaluation of the previous studies shows that they made a good attempt at explaining the

relationship between information asymmetry and borrowing cost. However, there exists a

contextual gap in the sense that there seems no study evaluated the effect of information

asymmetry on borrowing cost among MFI clients in Kenya. Most of MFI clients are generally

information opaque because they do not have prior borrowing record besides not keeping

business records. In addition, it should not be assumed that the results obtained in other

contexts can be generalized into the Kenyan context. This is as evidenced by Kysucky and

Norden (2013) who noted that the benefits of information availability in influencing lending cost

are more prominent in developed countries where bank competition is high as opposed to

developing countries.

A critical look at the previous studies also shows that they did not comprehensively

operationalize the study variables. For instance, most studies only used credit history as the

only indicator of information asymmetry. Majority of studies also relied on a single measure of

borrowing cost namely interest rates. This study measured the two variables in a broader sense.

Information asymmetry was measured using: credit history, soft information and borrower-lender

proximity. While borrowing cost was measured using: interest rates, collateral requirement, debt

maturity and credit rationing.



CONCEPTUAL FRAMEWORK AND VARIABLE OPERATIONALIZATION

The conceptual framework in Figure 1 illustrates the relationship among study variables:

In this study, the dependent variable is cost of borrowing. This variable was measured using:

interest rates, collateral requirement, debt maturity, and credit rationing. The independent

variable in this study was information asymmetry. This variable was measured using: credit

history, soft information, and borrower proximity.

Figure 1 Conceptual Framework

Information asymmetry Cost of borrowing

Operationalization of Variables

The following table shows how the study variables will be measured, measurement scale to be

used and the corresponding questions in the questionnaire.

Table 1 Operationalization of the study variables

Variable Variable type Indicators Measurement scale No. of Questions in

questionnaire

Information

asymmetry

Independent Credit history Interval/Ordinal Question 7

Independent Soft information Ordinal/Interval Question 8

Independent Borrower proximity Ordinal/Interval Question 9

Borrowing Cost Dependent Interest rate

Collateral

requirement

Credit rationing

Debt maturity

Ordinal/Interval Question 10

Credit history

Soft information

Borrower proximity

Interest rates

Collateral requirement

Debt maturity

Credit rationing

http://ijecm.co.uk/

© Muli & Muli


METHODOLOGY

Research Design

This study adopted a descriptive survey research design. A descriptive study is done so as to

be able to give a rundown of the attributes of the study‟s variables and answer the research

questions. According to Best and Khan (2009) a descriptive survey involves presentation of an

issue highlighting its state, customs, opinions, operations, associations or directions. It involves

collecting information about existing circumstances or situations with the objective of

explanation and exposition. Aggarwal (2008) adds that this research method is not simply about

assembling and arranging facts but involves proper examination, explanation, collation,

determination of directions and associations.

Consistent with the foregoing explanation about the meaning of a descriptive survey

design, the purpose of this study will be to analyse and identify relationships between

information asymmetry and borrowing costs among microfinance clients in different MFIs. This

justifies the choice of this design. A descriptive survey method was used to collect data from

MFIs operators registered with the Association of Microfinance Institutions (AMFI) and

subsequently analysed the same to make sense of relationships.

Target Population

This study sought to find out the effect of information asymmetry on the borrowing cost among

MFIs borrowers. Therefore, the target population of this study was all the microfinance

institutions operating in Kenya as captured in the Association of Microfinance Institutions (AMFI)

data base. Currently (the year 2017) we have 49 MFIs that are registered with AMFI. See

appendix 1 for the sample frame. These will form the total population of this study. The unit of

analysis was the individual microfinance institutions. One key informant respondent per MFI was

targeted. The respondent was the MFI‟s credit manager. This was chosen as the respondent

because s/he is in charge of appraising loan applicants and determining whether the application

should be accepted or declined. The officer also determines the loan terms in relation to interest

rate, collateral requirement and loan maturity.

Sample Size and Sampling Procedure

The sample size was determined using a mathematical model propounded by the Air University

of USA (2002). The model is specified as follows:

n = {NZ2 ×0.25}/ {[d2 × (N-1)] + [Z2 ×0.25]}

Where; n = sample size, N = sample frame, d = precision level, Z=number of standard deviation

units corresponding to confidence level, 1 = mathematical constant.



According to the authors, this model is appropriate for determining sample size if you plan to

report results in a variety of ways. Some formulas are relevant where the researcher wants to

report results descriptively while other formulas are applicable where the reporting will be done

using both descriptive and inferential statistics. In this study the results will be reported by use of

both descriptive and inferential statistics, hence choice of the above model. The total population

(N) is 49 and the researcher will wish for 95% confidence level and ± 5 percent precision level

(d = 0.05, Z= 1.96). Accordingly, the sample size shall be:

N = {49 × 1.962 × 0.25}/ {[0.052× (49-1)] + [1.962×0.25]} = 43.56 ͌ 44

This model has previously been applied in various studies to compute sample size, for

instance; Muli (2011) and Gupta et al. (2005). The researcher adopted simple random sampling

procedure to select the 44 sample from the population list focusing on only those MFIs that have

branches within Nairobi County for the researcher‟s convenience in terms of time and financial

constraints. Simple random sampling is appropriate if the elements constituting the population

are similar. In this study there was no need to stratify the population constituent elements

because the elements are considered uniform and hence the researcher intended to randomly

select 44 MFIs from the total population.

Research Instrument

A five point structured likert scale questionnaire was used for the purposes of data collection.

The point of scale indicates the degree of agreement level of the respondent to the statement

indicating the effects of information asymmetry on the cost of borrowing. The questionnaire

comprises questions related to credit history, soft information and borrower proximity, and how

these affects borrowing cost.

In addition, the questionnaire also has further questions that measures the following

demographic aspects: age of the MFI, size of the MFI as measured using number of branches,

type of the MFI – whether deposit taking or not, lending modality – whether individual or group

lending, whether the MFI uses credit reference bureau services or not, and whether the MFI has

a dedicated relationship banking section.

Validity and Reliability of the Instrument

Validity shows the extent to which a tool measures the construct being investigated (Gall. et al.,

2003). The researcher used intelligible language in constructing the data collection tools such

that the questions were understandable and simple for the respondents to comprehend and

offer appropriate responses. The researcher ensured that the questions asked spoke to the

objectives of the study.

http://ijecm.co.uk/

© Muli & Muli


The questionnaire was pilot-tested to guarantee that the instruments ensured coherence and

pertinence. This involved administering the questionnaire in the filed on a sample of MFIs before

the actual study. Any problems in the sample questionnaire were identified and rectified before

the actual study. The views of the respondents were considered in improving the tool.

Reliability indicates the extent of internal consistency or dependability of a tool over time.

It shows the exactness and meticulousness of the measurement technique (Kothari, 2004). In

this study, the reliability of the questionnaire was tested using the Cronbach alpha coefficient.

According to Gliem and Gliem (2003) Cronbach alpha is the best measure of internal

consistency of a measurement instrument. The cut-off point for making a conclusion about the

reliability of a tool is the alpha coefficient should be ≥ 0.70.

Data Collection Procedure

This study collected quantitative data using a self-administered questionnaire. The researcher

informed the respondents that the instruments being administered was for research purpose

only and the responses from the respondents would be kept secret and confidential.

The researcher obtained an introductory letter from the University to collect data from the MFIs

and with the help of research assistants deliver the questionnaires to the respondents and have

them filled and then collect them later: the drop and pick later method.

Data Analysis

Descriptive and inferential statistics were used to analyze the sample data to explain and make

abstractions of the population phenomenon. The collected data was checked for completeness,

edited, coded and then entered into the computer. Descriptive statistics were used to

summarize the data and establish characteristics of the study population. The tools of analysis

used were frequency distributions and percentages. Mean was computed to show the average

opinion consensus on every likert scale question asked, while standard deviation was calculated

to show the extent of divergence in the opinion among the respondents.

To establish the effect of information asymmetry on the cost of borrowing, a multiple

regression model was used. The variables are a mixture scale of ordinal and interval

measurements. The model is specified as follows:

CR= β0 +β1CH+ β2SI + β3BP+ε

Where:

CR= Dependent variable (Cost of borrowing)

β0= constant term

β1, β2, β3 and β4> 0 are regression coefficients



ε = Stochastic term.

CH = credit history

SI = soft information

BP = borrower proximity

The regression coefficients tested the unique effect of each independent variable. Significance

of the regression coefficients was tested using the P values. The corresponding probability

value (P value) for each coefficient was used to test the significance of regression coefficients at

5% significance level. P value measures the probability of committing type II error. It is the

lowest significance level at which the null hypothesis can be rejected.

The ANOVA F statistic test was used to test whether the model as a whole is significant.

In other words, do the independent variables, taken together, predict the dependent variable

better than just predicting the mean of everything? R squared shall be used to test explanatory

power of the model.

Regression diagnostics play a vital role in finding and validating a good predictive

relationship between the dependent and independent variables. Since this is a cross sectional

study only two diagnostic tests were performed namely normality test and multicollinearity test.

Histograms and measures of skewness and kurtosis were used to measure normality of the

data. Multicollinearity occurs when two or more predictors in the model are correlated and

provide redundant information about the response. Pearson correlation was used to test the

strength of correlation between the predictor variables. The rule of the thumb is that a

correlation of more than 0.7 is an indication of severe multicollinearity (Shen and Gao, 2008).

The solution for this problem is to drop the affected variable.

RESULTS AND DISCUSSIONS

Response Rate

The study targeted a sample size of 44 respondents from different microfinance institutions

(MFIs). The questionnaire was supposed to be filled by the credit manager/officer from each

MFI. In this study 32 dully filled questionnaires were received from the respondents. This

represents 73% of the targeted sample size. The response rate conforms to Mugenda and

Mugenda (2003) stipulation that a response rate of 70% and over is acceptable.

Reliability of the Data

The quality of data is as good as the tool used to collect the data. To achieve reliability of the

data, the researcher computed Cronbach‟s Alpha Coefficient. This statistic was used to

http://ijecm.co.uk/

© Muli & Muli


ascertain internal consistency of the Likert scale questionnaire used in data collection. The

general rule of thumb is that a coefficient of not less than 0.7 lends credence to the internal

consistency of items used to measure the construct (Gliem & Gliem, 2003). Alpha values were

estimated scale-wise (That‟s separately for credit history, soft information and borrower

proximity). The analysis resulted in alpha coefficients higher than 0.70 in all the three cases.

The results are as captured in table 2 below.

Construct validity was demonstrated by high correlations between the items that

comprised the constructs. This is a strong indication that the data collection tool used was

reliable. Cronbach alpha is a reliability coefficient based on the average covariance among

items in a scale. The average correlation of an item with all other items in the scale articulates

the extent of the common entity.

Table 2 Cronbach alpha coefficients

Reliability Statistics (credit history)

Cronbach's

Alpha

Cronbach's

Alpha Based on

Standardized

Items N of Items

.906 .925 4

Reliability Statistics (soft information)

Cronbach's

Alpha

Cronbach's

Alpha Based on

Standardized

Items N of Items

.798 .819 4

Reliability Statistics (borrower proximity)

Cronbach's

Alpha

Cronbach's

Alpha Based on

Standardized

Items N of Items

.945 .967 4



Demographic Information

The demographic information of interest in this study was organization type, age of the MFI,

number of branches, lending modalities, and the use of credit reference bureaus in credit policy.

These demographics are discussed below:

Organization Type

The results of the study showed that 53 percent of the randomly sampled MFIs were non-

deposit taking while 47 percent are microfinance banks (meaning that they are allowed to

mobilize deposits from the public). Traditionally, MFIs in Kenya have not been allowed to

mobilize savings from the public until the year 2006 when the microfinance Act was enacted.

These results show a trend of MFIs increasingly embracing deposit taking in light of the

enabling legislation. However, a big portion of MFIs are yet to embrace deposit taking. Public

savings in MFIs is usually the main source of funds for lending purposes. The results are

summarized in table below

Table 3 Organization type

Organization type

Frequency Percent Valid Percent Cumulative Percent

Valid Non-deposit taking MFI 17 53.1 53.1 53.1

Microfinance Banks 15 46.9 46.9 100.0

Total 32 100.0 100.0

Age of the MFI

The results of this study showed that approximately 69 percent of the MFIs in Kenya have not

been in existence for more than ten years. Only 31 percent of MFIs in Kenya have been in

existence for more than 10 years. The proliferation of MFIs in the last ten years can be possibly

explained the enactment of the Microfinance Act (2006) which was done roughly ten years back.

Table 4 MFIs Age Distribution

Age


Valid 1 – 5 9 28.1 28.1 28.1

6 – 10 13 40.6 40.6 68.8

Above 10 10 31.3 31.3 100.0

Total 32 100.0 100.0

http://ijecm.co.uk/

© Muli & Muli


MFI branch network

Most MFIs in Kenya are fairly small in size with 63 percent of them having less than 10

branches country wide. Only 22 percent of the MFIs have more than 10 branches countrywide.

This implies that a satisfactory MFI outreach in the country is yet to be attained. However, the

tremendous proliferation on new MFIs in the last 10 years is a good indication. The summary of

MFI branch coverage is outlined in table below:

Table 5 MFI branch network

Number of Branches

Frequency Percent Valid Percent

Cumulative

Percent

Valid Less than 10 20 62.5 62.5 62.5

10 – 20 5 15.6 15.6 78.1

More than 20 7 21.9 21.9 100.0

Total 32 100.0 100.0

MFI Lending Modalities

In this study a question was posed to the respondent as to whether the MFI lends to individual

borrowers, group of borrowers (group lending) or adopts both approaches. Approximately 56

percent of the MFIs use group lending modality, only 44 percent of the MFIs use individual

lending approach exclusively. Most MFIs adopt the group lending approach because it

minimizes default risk due to peer pressure among the group members (Table 6).

Table 6 MFI lending Modality

Lending modality


Cumulative

Percent

Valid Individual Lending 14 43.8 43.8 43.8

Group Lending 16 50.0 50.0 93.8

Both 2 6.3 6.3 100.0

Total 32 100.0 100.0

Use of Credit Reference Bureau by MFIs

The researcher sought to find out the extent to which MFIs have adopted the use of credit

reference bureaus in their lending framework. The results showed that 81 percent of MFIs in

Kenya use the services of credit reference bureaus in determining the borrower credit



worthiness. This shows the centrality of credit history in evaluating a loan applicant among MFIs

in Kenya. Table 7 below shows the results.

Table 7 MFI use of credit reference bureaus

MFI uses credit reference bureaus in lending policy


Cumulative

Percent

Valid No 6 18.8 18.8 18.8

Yes 26 81.3 81.3 100.0

Total 32 100.0 100.0

Descriptive Analysis

In this section, the study presents summary information about the opinion of respondents on

different constructs in the form of descriptive statistics. Generally, there were varied opinions on

the role of each independent variable in influencing the dependent variable.

Influence of credit history on the borrowing cost

Table 8a gives a summary of respondents‟ opinions on the effect of credit history on the

borrowing cost. The results show a divided opinion, while 37.5 percent of the respondents

agreed that a good credit history translates into relatively lower interest rates being charged, an

equivalent number of respondents held the contrary opinion (disagreed). The results also

indicated that 21.9 percent of the respondents opined that credit history moderately influences

the cost of borrowed. When the researcher lumps together those that agreed with those who

moderately agreed the cumulative percentage of this category is 59.4 percent. This implies that

credit history is an important determinant of the interest levied on borrower – that is, a better

credit history is likely to translate into lower interest rates.

Table 8a Effect of credit history on borrowing cost

Credit history


Valid Strongly Disagree 1 3.1 3.1 3.1

Disagree 12 37.5 37.5 40.6

Moderate 7 21.9 21.9 62.5

Agree 12 37.5 37.5 100.0

Total 32 100.0 100.0

http://ijecm.co.uk/

© Muli & Muli


The mean value of 2.9375 ≈ 3.000, as captured in table 8b below shows that the general

opinion on average was that credit history has an effect on borrowing cost to a moderate extent.

Note that the vale 3 corresponds to the opinion “Moderate”. However, the standard deviation

value of 0.94826 (not close to zero) is an indication that there was a diversity of opinion on

whether or not credit history affects borrowing cost and to what extent. Note that a standard

deviation value of zero would imply there was 100% consensus on the respondents‟ opinion.

Table 8b Mean and standard deviation

N Mean Std. Deviation

Statistic Statistic Statistic

Credit history 32 2.9375 .94826

Valid N (listwise) 32

Influence of soft information on borrowing cost

This study sought respondents‟ opinion on whether soft information has influence on the cost of

borrowing. In this respect there was also no consensus. Generally, 46.9 percent (21.9 + 21.9 +

3.1) of the respondents agreed that borrower soft information in possession of the MFI has an

influence on the cost of lending. Possession of positive soft information has an effect of reduced

borrowing cost and vice versa. On the other hand, 53.1 percent (15.6 + 37.5) of respondents

disagreed that soft information has an effect on the borrowing cost. This implies that there is

polarized opinion on this issue.

Table 9a Effect of soft information on borrowing cost

Soft information


Cumulative

Percent


Disagree 12 37.5 37.5 53.1

Moderate 7 21.9 21.9 75.0

Agree 7 21.9 21.9 96.9

Strongly Agree 1 3.1 3.1 100.0

Total 32 100.0 100.0

The researcher obtained further information by computing the mean and standard deviation of

the responses. As seen in table, the average consensus lied at 2.5937 ≈ 3.000. The

interpretation of this is that the respondents agreed to a moderate extent that soft information



has an effect on the borrowing cost. However, a standard deviation score of 1.10306 implies

that there was a polarized opinion on this issues, that is, the respondents held extreme opinions

on both ends. These measures of central tendency and dispersion are shown in table below.


Descriptive Statistics


Soft information 32 2.5937 1.10306


Effect of Borrower Proximity on borrowing cost

In this study it was hypothesized that borrower proximity has a bearing on the cost of borrowing.

In response to this question, 90.6 percent of the respondents disagreed that borrower proximity

affects borrowing cost. On the other hand, 9.4% agreed that borrower proximity affects

borrowing cost. These results are shown in table 10a below.

Table 10a Influence of borrower proximity on borrowing cost

Borrower proximity



Disagree 16 50.0 50.0 90.6

Moderate 1 3.1 3.1 93.8

Strongly Agree 2 6.3 6.3 100.0

Total 32 100.0 100.0

On average, the respondents disagreed that borrower proximity affects borrowing cost. This is

as evidenced by the mean score of 1.8125 ≈ 2.000. Note that the value 2 corresponds to the

opinion “Disagree”. Furthermore, a standard deviation value of 0.99798 (far from zero) shows

that there was a divided opinion on this question.


Descriptive Statistics


Borrower proximity 32 1.8125 .99798


http://ijecm.co.uk/

© Muli & Muli


Regression Model

An estimation of the relationship between the independent variables and the dependent variable

was conducted using a regression model. In this study, borrowing cost was the dependent

variable while the independent variables included credit history, soft information, and borrower

proximity.

Regression Coefficients

Table 11 Regression coefficients

Coefficientsa

Model

Unstandardized Coefficients

Standardized

Coefficients

t Sig. B Std. Error Beta

1 (Constant) -.887 .223 -3.977 .000

CH .703 .138 .469 5.095 .000

SI .720 .135 .586 5.338 .000

BP -.090 .112 -.063 -.799 .431

a. Dependent Variable: CR

The estimated regression equation takes the form: CR= -0.887 + 0.46CH+0.586SI – 0.063BP

The results show that credit history and soft information have a significant influence on the

borrowing cost among MFI borrowers. This is as supported by the P values of 0.000 for the

variables regression coefficients. However, borrower proximity has not relationship with the

borrowing cost. This is as evidenced by the regression coefficient‟s P value of 0.431 which is

way above the benchmark significance value of α = 0.05.

Both credit history and soft information have a favourable influence on the borrowing

cost. That is to imply that any positive information as regards credit history and soft information

leads to reduced borrowing cost. The positive regression coefficients of 0.469 and 0.586 for

credit history and soft information respectively attest to this. The reduced borrowing cost could

come in the form of lower interest rates, non-requirement of collateral, non-credit rationing and

longer repayment periods.

Model Explanatory Power

The regression model explanatory power was measured using R squared. The results are

shown below in table 12.



Table 12 Model summary

Model Summary

Model R

R

Square

Adjusted R

Square

Std. Error of

the Estimate

Change Statistics

R Square

Change F Change df1 df2

Sig. F

Change

1 .969a .938 .932 .37136 .938 142.265 3 28 .000

a. Predictors: (Constant), BP, CH, SI

The results show that 93.20% of the changes in the borrowing cost among the MFI borrowers is

explained by credit history and soft information. This as evidenced by the R squared value of

0.932 and a significant P value of 0.000.

Significance of the Overall Model

The significance of the overall model was tested using the F-statistic. The results indicate that

all the independent variables are a good joint predictor of the dependent variable. From the

following ANOVA table, the F-statistic value obtained was 142.265 which was significant at α =

0.05 as supported by the p-value of 0.000.

Table 13 One Way ANOVA

ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression 58.857 3 19.619 142.265 .000b

Residual 3.861 28 .138

Total 62.719 31

a. Dependent Variable: CR

b. Predictors: (Constant), BP, CH, SI

SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

Summary of the Findings

Money and credit are the lifeblood of an economy. The ability of borrowers to access credit at

reasonable terms is critical to facilitate investment and commerce, and thereby sustain

economic growth. Most microfinance clients are information opaque and this partly explains the

high risk premium attached to lending hence higher interest rates. Therefore, the poor still find it

difficult to access finances from MFIs because of the fairly high cost of loans. MFIs usually take

different measures aimed at protecting the lender against information asymmetry namely higher

interest, collateral requirement, shorter debt maturity and credit rationing which constitute a cost

http://ijecm.co.uk/

© Muli & Muli


to the borrower. This study sought to find out the effect of information asymmetry on the cost of

borrowing among the microfinance clients in Kenya.

The first objective of the study was to find out the effect of credit history on the cost of

borrowing among MFI clients in Kenya. The regression equation model results revealed that

credit history has a significant effect on the borrowing cost among the MFI clients. The

regression analysis returned a positive coefficient of + 0.46 (P value = 0.000). This means that a

good credit history has the effect of accessing credit at lower borrowing cost among the MFIs

borrowers in Kenya. These findings are as corroborated by Barbosa and Marcal (2011), Ngugi

(2001), Howorth and Moro (2012), and Huang et al. (2014).

The second objective of the study was to find out the effect of soft information on the

cost of borrowing among microfinance borrowers in Kenya. The analysis showed that soft

information has a significant effect on the cost of borrowing among the microfinance borrowers

in Kenya. The regression results returned a positive coefficient of +0.586 (P value = 0.000). This

implies that favorable soft information has the effect of lowering the borrowing cost among the

MFI borrowers in Kenya. These results are similar to those obtained by Campbell and Loumioto

(2013), Wanzare (2010), Aurizion, et al. (2012), and Sseggujja (2010). However, Kysucky and

Norden (2013) after conducting a meta-analysis concluded that higher prevalence of

relationship lending (soft information) does not necessarily come along with higher benefits for

borrowers.

The third objective of the study was to find out the effect of borrower proximity on the

cost of borrowing among microfinance borrowers in Kenya. The regression results showed that

borrower proximity has no relationship with the cost of borrowing among MFI borrowers in

Kenya. The analysis returned a P value of 0.431 which is way above the benchmark α = 0.05

significance level. However, this is contrary to results obtained in some other contexts like by

Bellucci et al. (2013) and Ostromogolsky (2017).

The results further showed that credit history and soft information are good joint

predictors of the borrowing cost among microfinance clients in Kenya. This is as supported by

the significant F statistic value of 142.265 (P value = 0.000). The results indicated that holding

other factors constant 93.20% of changes in the cost of borrowing among MFI clients in Kenya

is explained by credit history and soft information.

Conclusion

The study findings show that borrower credit history and soft information are significant

influencers of borrowing cost among microfinance borrowers in Kenya. Favorable borrower

credit history and soft information have a favorable effect of reducing the borrower cost of



borrowing among microfinance borrowers in Kenya. The reduced borrowing cost could come in

different ways which includes: lower interest rates, collateral requirement waiver, longer loan

maturity and non-rationing of the credit.

The study also concludes that in the Kenyan context borrower proximity has no effect on

the cost of borrowing among microfinance borrowers in Kenya. This implies that microfinance

institutions in Kenya do not take into account the physical distance of the borrower from the MFI

when evaluating a loan application. Therefore, the MFI will only consider other factors in making

the lending decision but not borrower proximity.

From the theoretical perspective, the findings of this study concur with the information

asymmetry theory argument, that information asymmetry can lead to an increase in borrowing

cost. This theory as it relates to the credit market was originally advanced by Akerlof (1970). In

general, the price of a loan is based on the lender‟s cost plus a risk premium. The size of this

premium depends on the lender‟s ability to properly assess the creditworthiness of the borrower.

Presence of information asymmetry therefore implies that this premium will be high hence

higher borrowing cost.

Recommendations

The study recommends that microfinance institutions in Kenya should leverage borrower credit

history and soft information to mitigate information asymmetry challenges. This practice will also

be beneficial to the borrowers who possess good credit history standing and positive soft

information. By extension this will make the bottom of the society (microfinance clients) to

access credit at favorable terms. In turn, this will have a positive impact on the economy by

mainstreaming the poor into economic participation through affordable credit which can be used

to finance business operations sustainably.

Credit reference bureaus (CRBs) play a very central role in bridging the information gap

between the borrowers and lenders. Therefore, the relevant policy makers should work on

growing the presence of more CRBs in the country. This will include availing a policy (legal)

environment that fosters the growth of CRBs and use of their services by the lending

institutions. Credit Reference Bureaus complement the central role played by banks and other

financial institutions in extending financial services within an economy. CRBs help lenders make

faster and more accurate credit decisions. They collect, manage and disseminate customer

information to lenders with in a provided regulatory framework. Credit histories not only provide

necessary input for credit underwriting, but also allow borrowers to take their credit history from

one financial institution to another, thereby making lending markets more competitive and, in the

end, more affordable. Credit bureaus assist in making credit accessible to more people, and

http://ijecm.co.uk/

© Muli & Muli


enabling lenders and businesses reduce risk and fraud. Sharing of information between

financial institutions in respect of customer credit behaviour, therefore, has a positive economic

impact.

The study notes that MFIs should also take advantage of soft information to lower the

cost of borrowing to the good clients. Soft information is typically obtained through relationship

banking. One of the most powerful technologies available to reduce information problems in

small firm finance is “relationship lending.” Under relationship lending, banks acquire information

over time through contact with the firm, its owner, and its local community on a variety of

dimensions and use this information in their decisions about the availability and terms of credit

to the firm.

The focus of this study was to find out the effect of microfinance on borrowing cost

among the microfinance borrowers in Kenya. The study only zeroed in on the MFIs operating

within Nairobi County which is an urban setting. There might be some issues that uniquely affect

the rural setting MFIs and clients. Therefore, it is recommended that a study that captures

clients in both urban and rural setting should be undertaken.

This was a cross-sectional study. Probably a time series study would be able to reveal

better information by tracking the borrowing cost of a client over time as s/he progressively

builds relationships with the lender as well as the lender compiles the customer credit history

over time.

REFERENCES

Agarwal, S., Chomsisengphet, S. Liu, C. and Souleles, N. (2009). Relationship Lending: Evidence from the Consumer Credit Market,” mimeo, Wharton.

Air University. (2002). Air University Sampling and surveying handbook. Air University, USA

Akerlof, G.A. (1970), “The Market for Lemons”, Quarterly Journal of Economics 84, 488-

500.

Amir, S. (2007). Information Asymmetry and Financing Arrangements: Evidence from Syndicated Loans, The Journal of finance, 1 (7)2, 200 – 210

Anderson, R.G. (2005).Monetary trends. Federal Reserve Bank of St. Louis. Available at www.research.stlouisfed.org

Aurizion, L.D., Oliviero, T. and Romano, L. (2012).Family firms and the agency cost of debt: the role of soft information during a crisis. EUI Working paper ECO 2012/22, European University Institute

Barbosa, R.C.O. and Marcal, E.F. (2011). The impacts of information asymmetry in determining bank spreads. Revista Gestao and Politicas Publicas, 1(2):113-130

Bellucci, A., Borisov, A. and Zazzaro, A. (2013). Do banks price discriminate spatially? Evidence from small business lending in local credit markets. CASMEF Working Paper Series, Working paper No. 3 (2013)

Belsley, D.A., Edwin, K & Roy, E.W. (1980). Regression diagnostics: Identifying influential data and sources of collinearity. New York: John Wiley and Sons

Best, J.W. & Kahn, J.V. (2007). Research in education, New Delhi, Prentice Hall of India Private

Bolton, P., Freixas, X., Gambacorta, L. and Mistrulli, P.O. (2013), Relationship and Transaction Lending in a Crisis, BIS Working Papers No 417



Bonfim, D., Dai, O. and Franco, F. (2009). The number of bank relationships, borrowing costs and bank competition, Banco de Portugal, Working Papers No. 12

Campbell, D. and Loumioto, M. (2013). Monitoring and the portability of soft information. Harvard Business School, Working paper 13 -077

Chodechai, S. (2004). Determinants of lending in Thailand: An empirical examination for the years 1992 – 1996, unpublished thesis

Dass, D. and Massa, M. (2009).The impact of a strong bank-firm relationship on the borrowing firm. Oxford university press, London

Emilio,C., Galariotis, Villa,C. & Yusupov, N. (2011).Recent advances in lending to the poor with asymmetric information. Journal of Development Studies, Taylor & Francis (Routledge), pp.1

Field, A. (2009). Discovering Statistics Using SPSS: Introducing Statistical Method (3rd ed.). Thousand Oaks, CA: Sage Publications.

Gaitho, N.W. (2013). Role of credit reference bureaus on credit access in Kenya: a survey of commercial banks in Kenya, European Scientific Journal, 9 (13)

George, D. and Mallery, P. ( 2010). SPSS for Windows Step by Step: A Simple Guide and Reference 17.0 Update. 10th Edition, Pearson, Boston.

Gliem, J.A. and Gliem, R.R. (2003). Midwest Research-to-practice conference in adult, continuing and community education. The Ohio State University, OH, October 8-10, 2003

Atul Gupta, S. Kanthi Herath, Nathalie C. Mikouiza, (2005) "Outsourcing in higher education: an empirical examination", International Journal of Educational Management, Vol. 19 Issue: 5, pp.396-412, https://doi.org/10.1108/09513540510607734

Hercules, S. (2006). Microcredit A Way to Overcome Problems of Financial Markets in Developing Countries? Lund University School of Economics and Management Department of Economics April 2006, Unpublished Masters Thesis,

Howorth, C. and Moro, A. (2012). Trustworthiness and interest rates: an empirical study of Italian SMEs. Small Business Economics, 39(1) 161-177

Huang, C.L. (2014). The Impact of Information Asymmetry and Client Credit on Lending Performance - Taiwan‟s Evidence, Asian Journal of Finance and Accounting, ISSN 1946-052X, Vol. 6, No.1

Huang, C., When, Y. and Liu, Z. (2014). Analysis on financing difficulties for SMEs due to asymmetric information. Global disclosure of economics and business, 3(2) ISSN 2305-9168

Institute of certified public accounts of Kenya - ICPAK. (2016).Banking sector reforms vital for growth. ICPAK, Nairobi

Jaap, B. Ralph, D.H. & Matteo, M. (2015). Sharing borrower information in a competitive credit market. The European Bank for Reconstruction and Development. Working Paper No. 180

Jappelli, T. and Pagano,M. (2002). Information Sharing, Lending and Defaults: Cross-Country Evidence”, Journal of Banking and Finance 26, 2017–45.

Jonathan, L. and Jonathan, T. (2001). Information asymmetry, small firm finance and the role of government. Journal of finance and management in public services (1), 43 – 60

Kothari C. R., (2004). Research Methodology: Methods and Techniques. New Delhi: New Age International (P) Ltd, Publishers.

Kwambai, K.D. and Wandera, M. (2013). Effects of credit information sharing on nonperforming loans: the case of Kenya commercial bank Kenya, European Scientific Journal, 9 (13) ISSN: 1857 – 7881 (Print) e - ISSN 1857- 7431

Kysucky, V. and Norden, L. (2013).the benefits of relationship lending in a cross- country context: a meta-analysis. Roterdam School of management, Erasmus University

Leitner, Y. (2006). Using Collateral to Secure Loans, Business Review Q2 www.philadelphiafed.org

Mehrteab, H. T. (2005). Adverse selection and moral hazard in group-based lending: evidence from Eritrea s.n. University of Groningen, thesis

Moerman, R.W. (2010).The Impact of Information Asymmetry on Debt Pricing and Maturity, The University of Chicago Booth School of Business

Mugenda A. and Mugenda (1999). Research Methods: Quantitative and Qualitative Approaches. Nairobi: Act Press

http://ijecm.co.uk/

© Muli & Muli


Muli, W.M. (2013). Financial Innovation as a Competitive Strategy: The Kenyan Financial Sector, Journal of Modern Accounting and Auditing, 9(7) 997-1004 http://www.davidpublishing.org/show.html?13182

Ngetich, J.C. and Wanjau, K. (2011). The Effects of interest rate spread on the level of non-performing assets: A case of commercial banks in Kenya. International journal of business and management, 1(1): 58-65

Ngugi, R.W. (2001). An empirical analysis of interest rate spread in Kenya. African Economic Research Consortium, AERC Research Paper 106

Ostromogolsky, P. (2017). Lender- borrower relationships and loan origination costs. Working Paper series, federal deposit Insurance Corporation (center for financial research), FDCI CFR WP 2017-03

Pandey I.M. (2005): Financial management, 9th Ed. Vikas Publishing House, India

Sabana B. (2003). Institutional assessment of micro-finance service providers in Kenya, DFID, Nairobi

Salvatore C, Stefano M. and Gabriele S. (2015). The how and why of soft information production in bank lending, Discussion paper series, Discussion Paper No. 10 Giugno 2015 - ISSN: 2280-9767, University of Naples Parthenope

Shen & Gao. (2008). A solution to Separation and Multicollinearity in Multiple Logistic Regression”, Journal of Data Science 6(2008), 515-53

Sseggujja, H.B. (2010).Relationship lending, transaction costs and the lending interest rates of commercial banks in Uganda. Unpublished MSc. Thesis, Makerere University

Stiglitz, J. E.and Weiss.A. (1981). “Credit Rationing in Markets with Imperfect Information,” American Economic Review, 71, 393-410.

Wanzare, N. (2010).The impact of hard and soft information on mortgage default. Unpublished M.A. thesis, University of Technology Sydney

effect of information asymmetry on borrowing cost …

Documents